You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
rocksdb/db/flush_job.cc

308 lines
11 KiB

// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
//
// Copyright (c) 2011 The LevelDB Authors. All rights reserved.
// Use of this source code is governed by a BSD-style license that can be
// found in the LICENSE file. See the AUTHORS file for names of contributors.
#include "db/flush_job.h"
#ifndef __STDC_FORMAT_MACROS
#define __STDC_FORMAT_MACROS
#endif
#include <inttypes.h>
#include <algorithm>
#include <vector>
#include "db/builder.h"
#include "db/db_iter.h"
#include "db/dbformat.h"
#include "db/event_helpers.h"
#include "db/filename.h"
#include "db/log_reader.h"
#include "db/log_writer.h"
#include "db/memtable.h"
#include "db/memtable_list.h"
#include "db/merge_context.h"
#include "db/version_set.h"
#include "port/likely.h"
#include "port/port.h"
#include "rocksdb/db.h"
#include "rocksdb/env.h"
#include "rocksdb/statistics.h"
#include "rocksdb/status.h"
#include "rocksdb/table.h"
#include "table/block.h"
#include "table/block_based_table_factory.h"
#include "table/merger.h"
#include "table/table_builder.h"
#include "table/two_level_iterator.h"
#include "util/coding.h"
#include "util/event_logger.h"
#include "util/file_util.h"
#include "util/iostats_context_imp.h"
#include "util/log_buffer.h"
#include "util/logging.h"
#include "util/mutexlock.h"
#include "util/perf_context_imp.h"
#include "util/stop_watch.h"
#include "util/sync_point.h"
#include "util/thread_status_util.h"
namespace rocksdb {
FlushJob::FlushJob(const std::string& dbname, ColumnFamilyData* cfd,
const DBOptions& db_options,
const MutableCFOptions& mutable_cf_options,
const EnvOptions& env_options, VersionSet* versions,
InstrumentedMutex* db_mutex,
std::atomic<bool>* shutting_down,
std::vector<SequenceNumber> existing_snapshots,
SequenceNumber earliest_write_conflict_snapshot,
JobContext* job_context, LogBuffer* log_buffer,
Directory* db_directory, Directory* output_file_directory,
CompressionType output_compression, Statistics* stats,
EventLogger* event_logger)
: dbname_(dbname),
cfd_(cfd),
db_options_(db_options),
mutable_cf_options_(mutable_cf_options),
env_options_(env_options),
versions_(versions),
db_mutex_(db_mutex),
shutting_down_(shutting_down),
existing_snapshots_(std::move(existing_snapshots)),
earliest_write_conflict_snapshot_(earliest_write_conflict_snapshot),
job_context_(job_context),
log_buffer_(log_buffer),
db_directory_(db_directory),
output_file_directory_(output_file_directory),
output_compression_(output_compression),
stats_(stats),
Allow GetThreadList() to report operation stage. Summary: Allow GetThreadList() to report operation stage. Test Plan: ./thread_list_test ./db_bench --benchmarks=fillrandom --num=100000 --threads=40 \ --max_background_compactions=10 --max_background_flushes=3 \ --thread_status_per_interval=1000 --key_size=16 --value_size=1000 \ --num_column_families=10 export ROCKSDB_TESTS=ThreadStatus ./db_test Sample output ThreadID ThreadType cfName Operation OP_StartTime ElapsedTime Stage State 140116265861184 Low Pri 140116270055488 Low Pri 140116274249792 High Pri column_family_name_000005 Flush 2015/03/10-14:58:11 0 us FlushJob::WriteLevel0Table 140116400078912 Low Pri column_family_name_000004 Compaction 2015/03/10-14:58:11 0 us CompactionJob::FinishCompactionOutputFile 140116358135872 Low Pri column_family_name_000006 Compaction 2015/03/10-14:58:10 1 us CompactionJob::FinishCompactionOutputFile 140116341358656 Low Pri 140116295221312 High Pri default Flush 2015/03/10-14:58:11 0 us FlushJob::WriteLevel0Table 140116324581440 Low Pri column_family_name_000009 Compaction 2015/03/10-14:58:11 0 us CompactionJob::ProcessKeyValueCompaction 140116278444096 Low Pri 140116299415616 Low Pri column_family_name_000008 Compaction 2015/03/10-14:58:11 0 us CompactionJob::FinishCompactionOutputFile 140116291027008 High Pri column_family_name_000001 Flush 2015/03/10-14:58:11 0 us FlushJob::WriteLevel0Table 140116286832704 Low Pri column_family_name_000002 Compaction 2015/03/10-14:58:11 0 us CompactionJob::FinishCompactionOutputFile 140116282638400 Low Pri Reviewers: rven, igor, sdong Reviewed By: sdong Subscribers: dhruba, leveldb Differential Revision: https://reviews.facebook.net/D34683
10 years ago
event_logger_(event_logger) {
// Update the thread status to indicate flush.
ReportStartedFlush();
Allow GetThreadList() to report operation stage. Summary: Allow GetThreadList() to report operation stage. Test Plan: ./thread_list_test ./db_bench --benchmarks=fillrandom --num=100000 --threads=40 \ --max_background_compactions=10 --max_background_flushes=3 \ --thread_status_per_interval=1000 --key_size=16 --value_size=1000 \ --num_column_families=10 export ROCKSDB_TESTS=ThreadStatus ./db_test Sample output ThreadID ThreadType cfName Operation OP_StartTime ElapsedTime Stage State 140116265861184 Low Pri 140116270055488 Low Pri 140116274249792 High Pri column_family_name_000005 Flush 2015/03/10-14:58:11 0 us FlushJob::WriteLevel0Table 140116400078912 Low Pri column_family_name_000004 Compaction 2015/03/10-14:58:11 0 us CompactionJob::FinishCompactionOutputFile 140116358135872 Low Pri column_family_name_000006 Compaction 2015/03/10-14:58:10 1 us CompactionJob::FinishCompactionOutputFile 140116341358656 Low Pri 140116295221312 High Pri default Flush 2015/03/10-14:58:11 0 us FlushJob::WriteLevel0Table 140116324581440 Low Pri column_family_name_000009 Compaction 2015/03/10-14:58:11 0 us CompactionJob::ProcessKeyValueCompaction 140116278444096 Low Pri 140116299415616 Low Pri column_family_name_000008 Compaction 2015/03/10-14:58:11 0 us CompactionJob::FinishCompactionOutputFile 140116291027008 High Pri column_family_name_000001 Flush 2015/03/10-14:58:11 0 us FlushJob::WriteLevel0Table 140116286832704 Low Pri column_family_name_000002 Compaction 2015/03/10-14:58:11 0 us CompactionJob::FinishCompactionOutputFile 140116282638400 Low Pri Reviewers: rven, igor, sdong Reviewed By: sdong Subscribers: dhruba, leveldb Differential Revision: https://reviews.facebook.net/D34683
10 years ago
TEST_SYNC_POINT("FlushJob::FlushJob()");
}
FlushJob::~FlushJob() {
ThreadStatusUtil::ResetThreadStatus();
}
void FlushJob::ReportStartedFlush() {
ThreadStatusUtil::SetColumnFamily(cfd_, cfd_->ioptions()->env,
cfd_->options()->enable_thread_tracking);
ThreadStatusUtil::SetThreadOperation(ThreadStatus::OP_FLUSH);
ThreadStatusUtil::SetThreadOperationProperty(
ThreadStatus::COMPACTION_JOB_ID,
job_context_->job_id);
IOSTATS_RESET(bytes_written);
}
void FlushJob::ReportFlushInputSize(const autovector<MemTable*>& mems) {
uint64_t input_size = 0;
for (auto* mem : mems) {
input_size += mem->ApproximateMemoryUsage();
}
ThreadStatusUtil::IncreaseThreadOperationProperty(
ThreadStatus::FLUSH_BYTES_MEMTABLES,
input_size);
}
void FlushJob::RecordFlushIOStats() {
ThreadStatusUtil::SetThreadOperationProperty(
ThreadStatus::FLUSH_BYTES_WRITTEN, IOSTATS(bytes_written));
}
Status FlushJob::Run(FileMetaData* file_meta) {
Allow GetThreadList() to report operation stage. Summary: Allow GetThreadList() to report operation stage. Test Plan: ./thread_list_test ./db_bench --benchmarks=fillrandom --num=100000 --threads=40 \ --max_background_compactions=10 --max_background_flushes=3 \ --thread_status_per_interval=1000 --key_size=16 --value_size=1000 \ --num_column_families=10 export ROCKSDB_TESTS=ThreadStatus ./db_test Sample output ThreadID ThreadType cfName Operation OP_StartTime ElapsedTime Stage State 140116265861184 Low Pri 140116270055488 Low Pri 140116274249792 High Pri column_family_name_000005 Flush 2015/03/10-14:58:11 0 us FlushJob::WriteLevel0Table 140116400078912 Low Pri column_family_name_000004 Compaction 2015/03/10-14:58:11 0 us CompactionJob::FinishCompactionOutputFile 140116358135872 Low Pri column_family_name_000006 Compaction 2015/03/10-14:58:10 1 us CompactionJob::FinishCompactionOutputFile 140116341358656 Low Pri 140116295221312 High Pri default Flush 2015/03/10-14:58:11 0 us FlushJob::WriteLevel0Table 140116324581440 Low Pri column_family_name_000009 Compaction 2015/03/10-14:58:11 0 us CompactionJob::ProcessKeyValueCompaction 140116278444096 Low Pri 140116299415616 Low Pri column_family_name_000008 Compaction 2015/03/10-14:58:11 0 us CompactionJob::FinishCompactionOutputFile 140116291027008 High Pri column_family_name_000001 Flush 2015/03/10-14:58:11 0 us FlushJob::WriteLevel0Table 140116286832704 Low Pri column_family_name_000002 Compaction 2015/03/10-14:58:11 0 us CompactionJob::FinishCompactionOutputFile 140116282638400 Low Pri Reviewers: rven, igor, sdong Reviewed By: sdong Subscribers: dhruba, leveldb Differential Revision: https://reviews.facebook.net/D34683
10 years ago
AutoThreadOperationStageUpdater stage_run(
ThreadStatus::STAGE_FLUSH_RUN);
// Save the contents of the earliest memtable as a new Table
FileMetaData meta;
autovector<MemTable*> mems;
cfd_->imm()->PickMemtablesToFlush(&mems);
if (mems.empty()) {
LogToBuffer(log_buffer_, "[%s] Nothing in memtable to flush",
cfd_->GetName().c_str());
return Status::OK();
}
ReportFlushInputSize(mems);
// entries mems are (implicitly) sorted in ascending order by their created
// time. We will use the first memtable's `edit` to keep the meta info for
// this flush.
MemTable* m = mems[0];
VersionEdit* edit = m->GetEdits();
edit->SetPrevLogNumber(0);
// SetLogNumber(log_num) indicates logs with number smaller than log_num
// will no longer be picked up for recovery.
edit->SetLogNumber(mems.back()->GetNextLogNumber());
edit->SetColumnFamily(cfd_->GetID());
// This will release and re-acquire the mutex.
Status s = WriteLevel0Table(mems, edit, &meta);
if (s.ok() &&
(shutting_down_->load(std::memory_order_acquire) || cfd_->IsDropped())) {
s = Status::ShutdownInProgress(
"Database shutdown or Column family drop during flush");
}
if (!s.ok()) {
cfd_->imm()->RollbackMemtableFlush(mems, meta.fd.GetNumber());
} else {
LogAndApply() should fail if the column family has been dropped Summary: This patch finally fixes the ColumnFamilyTest.ReadDroppedColumnFamily test. The test has been failing very sporadically and it was hard to repro. However, I managed to write a new tests that reproes the failure deterministically. Here's what happens: 1. We start the flush for the column family 2. We check if the column family was dropped here: https://github.com/facebook/rocksdb/blob/a3fc49bfddcdb1ff29409aacd06c04df56c7a1d7/db/flush_job.cc#L149 3. This check goes through, ends up in InstallMemtableFlushResults() and it goes into LogAndApply() 4. At about this time, we start dropping the column family. Dropping the column family process gets to LogAndApply() at about the same time as LogAndApply() from flush process 5. Drop column family goes through LogAndApply() first, marking the column family as dropped. 6. Flush process gets woken up and gets a chance to write to the MANIFEST. However, this is where it gets stuck: https://github.com/facebook/rocksdb/blob/a3fc49bfddcdb1ff29409aacd06c04df56c7a1d7/db/version_set.cc#L1975 7. We see that the column family was dropped, so there is no need to write to the MANIFEST. We return OK. 8. Flush gets OK back from LogAndApply() and it deletes the memtable, thinking that the data is now safely persisted to sst file. The fix is pretty simple. Instead of OK, we return ShutdownInProgress. This is not really true, but we have been using this status code to also mean "this operation was canceled because the column family has been dropped". The fix is only one LOC. All other code is related to tests. I added a new test that reproes the failure. I also moved SleepingBackgroundTask to util/testutil.h (because I needed it in column_family_test for my new test). There's plenty of other places where we reimplement SleepingBackgroundTask, but I'll address that in a separate commit. Test Plan: 1. new test 2. make check 3. Make sure the ColumnFamilyTest.ReadDroppedColumnFamily doesn't fail on Travis: https://travis-ci.org/facebook/rocksdb/jobs/79952386 Reviewers: yhchiang, anthony, IslamAbdelRahman, kradhakrishnan, rven, sdong Reviewed By: sdong Subscribers: dhruba, leveldb Differential Revision: https://reviews.facebook.net/D46773
9 years ago
TEST_SYNC_POINT("FlushJob::InstallResults");
// Replace immutable memtable with the generated Table
s = cfd_->imm()->InstallMemtableFlushResults(
cfd_, mutable_cf_options_, mems, versions_, db_mutex_,
meta.fd.GetNumber(), &job_context_->memtables_to_free, db_directory_,
log_buffer_);
}
if (s.ok() && file_meta != nullptr) {
*file_meta = meta;
}
RecordFlushIOStats();
auto stream = event_logger_->LogToBuffer(log_buffer_);
stream << "job" << job_context_->job_id << "event"
<< "flush_finished";
stream << "lsm_state";
stream.StartArray();
auto vstorage = cfd_->current()->storage_info();
for (int level = 0; level < vstorage->num_levels(); ++level) {
stream << vstorage->NumLevelFiles(level);
}
stream.EndArray();
return s;
}
Status FlushJob::WriteLevel0Table(const autovector<MemTable*>& mems,
VersionEdit* edit, FileMetaData* meta) {
Allow GetThreadList() to report operation stage. Summary: Allow GetThreadList() to report operation stage. Test Plan: ./thread_list_test ./db_bench --benchmarks=fillrandom --num=100000 --threads=40 \ --max_background_compactions=10 --max_background_flushes=3 \ --thread_status_per_interval=1000 --key_size=16 --value_size=1000 \ --num_column_families=10 export ROCKSDB_TESTS=ThreadStatus ./db_test Sample output ThreadID ThreadType cfName Operation OP_StartTime ElapsedTime Stage State 140116265861184 Low Pri 140116270055488 Low Pri 140116274249792 High Pri column_family_name_000005 Flush 2015/03/10-14:58:11 0 us FlushJob::WriteLevel0Table 140116400078912 Low Pri column_family_name_000004 Compaction 2015/03/10-14:58:11 0 us CompactionJob::FinishCompactionOutputFile 140116358135872 Low Pri column_family_name_000006 Compaction 2015/03/10-14:58:10 1 us CompactionJob::FinishCompactionOutputFile 140116341358656 Low Pri 140116295221312 High Pri default Flush 2015/03/10-14:58:11 0 us FlushJob::WriteLevel0Table 140116324581440 Low Pri column_family_name_000009 Compaction 2015/03/10-14:58:11 0 us CompactionJob::ProcessKeyValueCompaction 140116278444096 Low Pri 140116299415616 Low Pri column_family_name_000008 Compaction 2015/03/10-14:58:11 0 us CompactionJob::FinishCompactionOutputFile 140116291027008 High Pri column_family_name_000001 Flush 2015/03/10-14:58:11 0 us FlushJob::WriteLevel0Table 140116286832704 Low Pri column_family_name_000002 Compaction 2015/03/10-14:58:11 0 us CompactionJob::FinishCompactionOutputFile 140116282638400 Low Pri Reviewers: rven, igor, sdong Reviewed By: sdong Subscribers: dhruba, leveldb Differential Revision: https://reviews.facebook.net/D34683
10 years ago
AutoThreadOperationStageUpdater stage_updater(
ThreadStatus::STAGE_FLUSH_WRITE_L0);
db_mutex_->AssertHeld();
const uint64_t start_micros = db_options_.env->NowMicros();
// path 0 for level 0 file.
meta->fd = FileDescriptor(versions_->NewFileNumber(), 0, 0);
Version* base = cfd_->current();
base->Ref(); // it is likely that we do not need this reference
Status s;
{
db_mutex_->Unlock();
if (log_buffer_) {
log_buffer_->FlushBufferToLog();
}
std::vector<InternalIterator*> memtables;
ReadOptions ro;
ro.total_order_seek = true;
Arena arena;
uint64_t total_num_entries = 0, total_num_deletes = 0;
size_t total_memory_usage = 0;
for (MemTable* m : mems) {
Log(InfoLogLevel::INFO_LEVEL, db_options_.info_log,
"[%s] [JOB %d] Flushing memtable with next log file: %" PRIu64 "\n",
cfd_->GetName().c_str(), job_context_->job_id, m->GetNextLogNumber());
memtables.push_back(m->NewIterator(ro, &arena));
total_num_entries += m->num_entries();
total_num_deletes += m->num_deletes();
total_memory_usage += m->ApproximateMemoryUsage();
}
event_logger_->Log() << "job" << job_context_->job_id << "event"
<< "flush_started"
<< "num_memtables" << mems.size() << "num_entries"
<< total_num_entries << "num_deletes"
<< total_num_deletes << "memory_usage"
<< total_memory_usage;
TableFileCreationInfo info;
{
ScopedArenaIterator iter(
NewMergingIterator(&cfd_->internal_comparator(), &memtables[0],
static_cast<int>(memtables.size()), &arena));
Log(InfoLogLevel::INFO_LEVEL, db_options_.info_log,
"[%s] [JOB %d] Level-0 flush table #%" PRIu64 ": started",
cfd_->GetName().c_str(), job_context_->job_id, meta->fd.GetNumber());
TEST_SYNC_POINT_CALLBACK("FlushJob::WriteLevel0Table:output_compression",
&output_compression_);
s = BuildTable(
dbname_, db_options_.env, *cfd_->ioptions(), env_options_,
cfd_->table_cache(), iter.get(), meta, cfd_->internal_comparator(),
cfd_->int_tbl_prop_collector_factories(), cfd_->GetID(),
cfd_->GetName(), existing_snapshots_,
earliest_write_conflict_snapshot_, output_compression_,
cfd_->ioptions()->compression_opts,
mutable_cf_options_.paranoid_file_checks, cfd_->internal_stats(),
Env::IO_HIGH, &table_properties_, 0 /* level */);
info.table_properties = table_properties_;
LogFlush(db_options_.info_log);
}
Log(InfoLogLevel::INFO_LEVEL, db_options_.info_log,
"[%s] [JOB %d] Level-0 flush table #%" PRIu64 ": %" PRIu64
" bytes %s"
"%s",
cfd_->GetName().c_str(), job_context_->job_id, meta->fd.GetNumber(),
meta->fd.GetFileSize(), s.ToString().c_str(),
meta->marked_for_compaction ? " (needs compaction)" : "");
// output to event logger
if (s.ok()) {
info.db_name = dbname_;
info.cf_name = cfd_->GetName();
info.file_path = TableFileName(db_options_.db_paths,
meta->fd.GetNumber(),
meta->fd.GetPathId());
info.file_size = meta->fd.GetFileSize();
info.job_id = job_context_->job_id;
EventHelpers::LogAndNotifyTableFileCreation(
event_logger_, db_options_.listeners,
meta->fd, info);
TEST_SYNC_POINT("FlushJob::LogAndNotifyTableFileCreation()");
}
if (!db_options_.disableDataSync && output_file_directory_ != nullptr) {
output_file_directory_->Fsync();
}
TEST_SYNC_POINT("FlushJob::WriteLevel0Table");
db_mutex_->Lock();
}
base->Unref();
// re-acquire the most current version
base = cfd_->current();
// Note that if file_size is zero, the file has been deleted and
// should not be added to the manifest.
if (s.ok() && meta->fd.GetFileSize() > 0) {
// if we have more than 1 background thread, then we cannot
// insert files directly into higher levels because some other
// threads could be concurrently producing compacted files for
// that key range.
// Add file to L0
edit->AddFile(0 /* level */, meta->fd.GetNumber(), meta->fd.GetPathId(),
meta->fd.GetFileSize(), meta->smallest, meta->largest,
meta->smallest_seqno, meta->largest_seqno,
meta->marked_for_compaction);
}
InternalStats::CompactionStats stats(1);
stats.micros = db_options_.env->NowMicros() - start_micros;
stats.bytes_written = meta->fd.GetFileSize();
cfd_->internal_stats()->AddCompactionStats(0 /* level */, stats);
cfd_->internal_stats()->AddCFStats(InternalStats::BYTES_FLUSHED,
meta->fd.GetFileSize());
RecordTick(stats_, FLUSH_WRITE_BYTES, meta->fd.GetFileSize());
return s;
}
} // namespace rocksdb